Despite the NCI's mission to reduce the burden of cancer on survivors, little progress has been made toward reducing psychosocial morbidity for pediatric brain tumor survivors (PBTS). Social deficits, as well as medical and neuro-cognitive impairment, threaten both their short and long-term quality of life. In our research, classmates have described PBTS as socially withdrawn, victimized and excluded by peers, less liked, and having fewer friends than control classmates. Girls and children treated with radiation have the worst outcomes, but social impairment is not limited to them. These difficulties, particularly as reported by peers, are strongly predictive of emotional, behavioral, and academic difficulties later in life, and PBTS demonstrate more of these deficits as adults than any other subgroup of pediatric cancer survivors. Unfortunately, we lack proven interventions to help these children. Research has yet to identify specific, teachable skills that, when deficient, account for poor social outcomes among PTBS. Social-ecological factors may moderate social outcomes for PTBS and could be targeted by interventions, but have not been examined for this population. Thus, our long- term goal is to reduce morbidity for PBTS by developing tailored interventions for them. The objective of this proposal is to identify the nature and extent of social information processing deficits in PBTS and determine if they, along with social-ecological factors, account for risk for social difficulties. We will recruit 212 PBTS (ages 8-13, >1 year post-treatment) from 3 pediatric hospitals. Mixed methods (i.e., questionnaires, performance- based and analog measures, video-taped interactions with friends) and multiple reporting sources (peer, teacher, mother, father, child) will be used in schools and homes to assess social competence (i.e., social information processing skills, social behavior and interactions, social adjustment) and contextual resources (i.e. parent-child relationship and parenting behavior, friendship quality, and school environment) that account for social competence. Identical data will be obtained for one control classmate per PBTS, matched for gender, race, and age. The rationale is that identifying these factors will allow the selection and design of effective interventions by knowing what to target for which PBTS. Our central hypothesis is that social information processing skills will account for differences in social behavior and adjustment for PBTS, and resources in their families, friendships, and schools will buffer the impact of the tumor and treatment on social competence. The proposed research is significant because it will enable us to develop effective services to minimize long-term morbidity in PTBS.